Establishment of the provirus, a DNA copy of the viral genome integrated into the host cell chromosome is an obligatory step in retroviral replication. Stable integration into the human genome is one of the primary reasons for the persistence of the human immunodeficiency virus (HIV) infection, which leads to the acquired immunodeficiency syndrome (AIDS). Therefore, HIV integrase (IN), the enzyme orchestrating the insertion of the DNA replica of the viral genome into the cellular chromosomal DNA, is a potential target for antiretroviral therapy. The recent discovery of the new lead compounds able to protect cells against HIV infection by inhibiting integration raises hope that integrase inhibitors will be added to future combination cocktails (for reviews see Debyser et al., 2002; Miller and Hazuda, 2001; Pani and Marongiu, 2000; Pommier et al., 2000).
Mechanistically and structurally, retroviral integrases (reviewed in Asante-Appiah and Skalkca, 1997; Brown et al., 1987; Craigie, 2001; Hindmarsh and Leis, 1999) are similar to the well-studied prokaryotic Mu and Tn5 transposases and belong to a family of DNA strand transferases that catalyze DNA strand cleavage/ligation via direct trans-esterification. In the course of viral integration, HIV IN performs two enzymatic reactions: first is the removal of the 3′-GT dinucleotides from the long terminal repeats (LTRs) (i.e. the 3′-processing reaction). The second reaction is the insertion of the recessed 3′-viral DNA ends into the opposite strands of the target DNA wherein the 3′ hydroxyls of the processed LTR ends attack two phosphodiester bonds in the target DNA molecule (the strand-transfer reaction). Insertion of the two viral LTRs takes place in a coordinated fashion across the major groove of the target DNA. As a result, the integrated provirus is flanked by two 5-nucleotide gaps as well as two mismatched 5′-AC dinucleotides, which are probably repaired by cellular enzymes (Yoder and Bushman, 2000). HIV IN, like other retroviral integrases, is comprised of three structural domains. The amino terminal domain contains the HHCC zinc finger motif, coordinates Zn and appears to be important for formation of the active IN multimers. The central, catalytic or “core” domain (residues 50-212) contains the active site (i.e. the DDE motif composed of D64, D116 and E152) and is structurally very similar to the catalytic domains of Mu phage and Tn5 transposases. In the context of the full-length enzyme, the core domain is thought to make sequence-specific contacts with the DNA. The carboxy terminal fragment possesses unspecific DNA-binding activity and is critical for multimerization of the enzyme. All single domain fragments of IN are dimeric when produced separately; the full-length HIV IN as well as other retroviral integrases form multimers under the conditions of the in vitro enzymatic assays (Ellison et al., 1995; Engelman et al., 1993). Recombinant HIV and avian sarcoma virus (ASV) integrases have been shown to behave as mixtures of monomers, dimers and tetramers (Coleman et al., 1999; Deprez et al., 2000; Jenkins et al., 1996). The presence of octamers and larger complexes has been suggested by some studies (Lee et al., 1997; Leh et al., 2000). Virion-associated HIV IN was shown to be in an oligomeric form, whereby dimers and higher order multimers appeared to be stabilized by disulfides, although the complexes were not studied under native conditions (Petit et al., 1999). Earlier, IN derived from avian myoblastosis virus (AMV) particles was demonstrated to be multimeric, behaving at least as a dimer (Grandgenett et al., 1978).
The 3′-processing and strand-transfer reactions can be reproduced in vitro using recombinant enzyme preparations and DNA substrates that mimic LTR ends. In vivo retroviral integration is preceded by the assembly of a stable and compact preintegration complex (PIC), that contains a 9.5 kb linear DNA copy of the viral genome associated with viral and cellular proteins (Brown et al., 1987; Bukrinsky et al., 1993; Ellison et al., 1990; Farnet and Haseltine, 1991; Miller et al., 1997). Several cellular proteins have been suggested to play auxiliary roles during retroviral integration. Thus, the barrier-to-autointegration factor (BAF) has been reported to protect Moloney murine leukemia virus (MoMLV) PICs against suicidal self-integration (Lee and Craigie, 1998). Another cellular protein, HMG-I(Y) was found in HIV PICs and appeared to be essential for their integration activity in vitro (Farnet and Bushman, 1997; Miller et al., 1997). Conversely, BAF could substitute for HMG-I(Y) at least in vitro, partially restoring integration activity of salt-denatured HIV-1 PICs (Chen and Engelman, 1998). Yet, BAF remains to be shown to co-fractionate with retroviral PICs. Both BAF and HMG-I(Y) are small DNA-binding proteins able to bridge and deform DNA molecules (Reeves and Beckerbauer, 2001; Zheng et al., 2000). They are thought to play structural roles within retroviral PICs, possibly juxtaposing both LTRs. Similarly, Mu phage transposase as well as enzymatically-unrelated λ phage integrase require the DNA-bending host proteins IHF and/or HU to form committed synaptic complexes (Friedman, 1992; Mizuuchi, 1992). IHF and HU appear to substitute for each other in promoting formation of the active λ integrase-DNA complexes in vitro (Segall et al., 1994). Another potential co-factor for HIV integration, the integrase interactor 1 (Ini1 or hSNF5/BAF47), was originally discovered in a yeast two-hybrid screen for human proteins interacting with HIV-1 IN (Kalpana et al., 1994). Ini1 is a subunit of the 2 MDa SWI/SNF chromatin-remodeling complex (Wang et al., 1996). It has been proposed that Ini1 plays a role during retroviral replication by directing the PICs to open chromatin regions (Kalpana et al., 1994) or by modulating expression of the integrated provirus (Turelli et al., 2001). Recent studies demonstrated that GFP-tagged Ini1 was exported from the nuclei of infected cells and co-localized with incoming sub-viral particles (Turelli et al., 2001). Ini1 has also been reported to enhance the release of infectious HIV particles; detectable amounts of Ini1 have been shown encapsulated within virions (Yung et al., 2001).
In order to develop effective therapies for the treatment and prevention of retroviral virus infectious diseases, such as AIDS, elucidation of the mechanisms by which it replicates in the human host is essential. Therefore, it is apparent that there still exists a need in the art for the identification and characterization of the cofactors which participate in the retroviral integration process.
The cloning and characterization of the protein Lens Epithelial Cell derived Growth Factor (LEDGF) is described in PCT/US98/09801. The patent application describes the stimulating activity of LEDGF on the proliferation of different cell types, particularly those of epithelial character, such as lens epithelial cells. In general, methods for treating cancer, inducing wound healing, inducing or inhibiting cell-death, cell differentiation and cell proliferation are provided. Also antibodies against LEDGF are described.
The LEDGF protein has been described as the positive transcription cofactor PC4-interacting protein (Ge et al., 1998). Two alternatively spliced cDNA clones were isolated from a HeLa library coding for two proteins p52 (333 amino acids, as predicted from the cDNA) and p75 (530 aminoacids) sharing identical 325 amino terminal residues. The transcripts coding for p75 and p52 were detected in different cell types and tissues, p52 being most abundant in testis and p75 in thymus. Independently, a cDNA clone coding for a protein identical to p75 has been isolated from a lens epithelium cell library (Singh et al., 1999). Overexpression of the protein stimulated survival of diverse primary cells and cell lines and enhanced their resistance to oxidative and hyperthermic stress (hence LEDGF, “lens epithelium derived growth factor”). The same protein has also been identified as the DFS70 autoantigen, antibodies to which were found in some cases of atopic dermatitis, asthma and interstitial cystitis (Ochs et al., 2000). One case of acute myeloid leukemia with a chromosomal translocation resulting in a fusion of the NUP98 and LEDGF genes has been described (Ahuja et al., 2000).